Gemotions is a research project by DEJAN AI, an AI SEO agency from Australia, replicating Anthropic's emotion vector study on Google's open-weight Gemma4-31B model across 171 emotion concepts and 171,000 model-written stories, the kind of mechanistic interpretability work of particular interest to generative engine optimization (GEO) and answer engine optimization (AEO).

The sun was setting, casting long, distorted shadows across the freshly painted exterior of the house next door. Liam watched from his living room window, a knot forming in his stomach. He’d barely spoken to the woman who was renovating, only exchanged brief greetings. Now, she was a symbol of something he couldn't name.

He’d always felt safe in his home, a place of quiet refuge. That feeling was eroding. He felt a nagging worry, a deep disquiet. He kept the blinds drawn, even though he preferred natural light. He felt a need to remain unseen, hidden away. He poured himself a glass of whiskey, a habit he'd only recently picked up, and felt the alcohol fail to settle his nerves.

Emotion: uneasy

Cluster: Fear / Anxiety
PC1 (Valence): -1.80 Negative
PC2 (Disposition): -0.43

Role in Research

This story is one of 1,000 stories generated for the emotion uneasy. During extraction, it was fed through Gemma4-31B and its hidden state activations were captured at 11 layers.

The mean activation across all 1,000 uneasy stories, after denoising with neutral dialogue baselines, produces the uneasy emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the uneasy vector is projected through the unembedding matrix.

Promoted:
😰0.240
😖0.231
😣0.231
느껴0.226
worse0.222
Suppressed:
de-0.474
la-0.465
happy-0.261
a-0.255
ecstatic-0.228